Developing new portfolio strategies by aggregation
The authors propose a method to optimally combine N strategies having a given utility function. These strategies should be numerically computed, there is no requirement that strategies should have closed-form expression. It is used a nonparametric method to compute the optimal weights of combination. The proposed approach does not depend on distributional assumptions of assets returns. Empirical studies on real-world data are presented using three utility functions and a pool of five portfolio strategies. The presented method is computationally efficient.
- A well-conditioned estimator for large-dimensional covariance matrices
- Adaptive Regression by Mixing
- Aggregating regression procedures to improve performance
- Combining different procedures for adaptive regression
- Computing efficient frontiers using estimated parameters
- Empirical properties of asset returns: stylized facts and statistical issues
- Online portfolio selection: a survey
- Resampling methods for dependent data
- The Stationary Bootstrap
- Universal Portfolios
- Quantile-based portfolios: post-model-selection estimation with alternative specifications
- Dynamic large financial networks \textit{via} conditional expected shortfalls
- Bagged Pretested Portfolio Selection
- Mean-variance efficient large portfolios: a simple machine learning heuristic technique based on the two-fund separation theorem
- Mixed frequency data and portfolio selection: a novel approach integrating DEA with mixed frequency data sources
This page was built for publication: Developing new portfolio strategies by aggregation
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q827154)
